A Multi-Level Clustering Framework for Provincial EducationalFacility Equalization in Indonesia

Abstract

Equitable distribution of educational facilities is crucial for development planning, as regional disparities in facility availability can constrain access to education. This study identifies priority areas for school-facility equalization in Indonesia based on 2024 data covering 38 provinces and village/urban-ward-based facility availability by education level. Unlike single-stage clustering studies, this study combines macro-prioritization and micro-level need profiling to identify which provinces should be prioritized but also which education levels require attention. The analysis includes log1p transformation, standardization, optimal cluster selection using Elbow and Silhouette criteria, and the application of Level-1 and Level-2 K-Means clustering. The Level-1 results produce three priority groups: High Priority, Medium Priority, and Low Priority, with the optimal structure at K=3. The Level-2 analysis within the high-priority group is most stable at K2=2, distinguishing provinces dominated by primary and lower-secondary facility shares from those with a more balanced composition and relatively higher tertiary share. The Silhouette values indicate that the selected clusters provide reasonably separated groupings. The proposed framework provides a data-driven priority map and level-specific need profiles. The results can support staged infrastructure planning and differentiated interventions across provincial priority groups to strengthen educational facility equalization in Indonesia.